To restore sight to ~150,000 veterans (and others) that are blind as the result of retinal degenerative diseases (e.g. macular degeneration or retinitis pigmentosa), several worldwide groups are developing a retinal prosthetic - a device designed to restore vision by electrically stimulating inner retinal neurons, large numbers of which have been shown to survive the degeneration process. Ongoing clinical testing provides strong evidence that such devices have the potential to restore high levels of vision to the blind e.g. they enable shape recognition, improved ambulation and even allow some limited reading. Recent regulatory approval allows commercial sale of two of these devices, making this approach a viable treatment option. Unfortunately however, clinical results have been inconsistent and the quality of elicited percepts still remains crude in most patients. Because the quality of elicited vision is thought to arise directly from the pattern of neural activity elicited in the retina, we study and compare the patterns of neural activity that arise from different forms of stimulation with the goal of developing more effective stimulation methods. Targeting bipolar cells (retinal interneurons) with electric stimulation is thought to generally be advantageous because it utilizes the surviving retinal circuitry to create neural activity that presumably is a closer match to normal (physiological) signaling patterns, and therefore, easier for the brain to interpret. While most existing devices are thought to target bipolar cells, the actual benefits of this approach have not been evaluated previously and will be the focus of the first aim. Retinal degeneration can alter the synaptic pathways that shape bipolar-mediated patterns and we will therefore also evaluate how efficacy changes during the course of degeneration. Additional work from our lab suggests that the fading of percepts reported during clinical use may be tied to a loss of sensitivity that arises in response to repetitive activation of the retinal network (i.e. from bipolar cell activatin). We have new preliminary results that show considerable variability in the amount of loss and suggest that some of the cell types thought to mediate conscious vision may be the ones most sensitive to repetitive stimulation. Our goal in Aim 2 is to determine the sensitivity of all key cll types and evaluate whether the loss in sensitivity can be mitigated by alternative (repetitive) stimulation schemes. The final aim will look at the effectiveness with which small electrodes create physio- logical signaling patterns. Our preliminary results suggest small electrodes are less effective in activating the network and we will study their effectiveness in both the healthy and degenerate retinas. To maximize the translational value of our work, all testing will be performed with a clone of one of the commercially available devices - the Alpha-IMS from Retina Implant, AG.